Insiders Reveal Major Problems at Lab-Grown-Meat Startup Upside Foods

On July 1, five diners sat down at the counter of the Michelin-starred Bar Crenn in San Francisco for an unusual meal. They had won a competition to become the first customers in the US to eat cultivated meat—real animal cells grown in bioreactors instead of a living animal. For a nominal price of $1, they tucked into two pieces of a cultivated chicken fillet made by the California startup Upside Foods, one of only two companies cleared to sell cultivated meat in the US. “I thought it was delicious,” says Oscar Merino, one of the diners. “The taste and the texture was incredible.”

Before their meal at Bar Crenn, Merino and the rest of the group toured Upside’s production facility across the Bay in Emeryville. They looked at neat rows of gleaming steel bioreactors, each one surrounded by a web of pipework. This factory—which WIRED visited in May 2022—is where Upside says it brews its cultivated chicken. The facility, Upside says, demonstrates to the world exactly how this novel meat is made. “We’re starting to show, from day one, what this whole industry is about,” Upside Foods cofounder and CEO Uma Valeti said in May 2022. “This is the opposite of very closely guarded food innovations.”

But former and current employees say the Emeryville plant tells a misleading story of how Upside’s chicken is made. In fact, sources say, the company’s flagship product—the juicy whole cuts of chicken served at Bar Crenn—are brewed, almost by hand, in tiny bottles. The huge bioreactors, those sources claim, simply aren’t capable of reliably brewing the sheets of tissue needed to form whole cuts of meat such as chicken fillets.

Insiders say that Upside’s meticulously crafted fillets are instead the result of a process that is more arduous and unwieldy than using bioreactors: Employees grow thin sheets of tissue in small plastic flasks called roller bottles and combine them to create a larger hunk of chicken, an approach that is expensive and requires many hours of labor to produce even a small amount of meat. According to former and current employees at Upside, this process happens in a laboratory that doesn’t feature in the factory tours Upside gives to journalists and members of the public.

Faced with the considerable challenge of growing whole-cut meat at scale, most cultivated meat companies have decided to focus on the more modest goal of brewing cells using cheaper, more established bioreactor technology more suited to producing ground products like chicken nuggets and burgers. Upside, which is widely viewed as the leading company in the space, has long attracted attention by implying that it is ready to produce whole cuts of chicken at large volumes, a breakthrough that, if true, would put it far ahead of the competition.

Bioreactors used for brewing cultivated meat. Photograph: Christie Hemm Klok

In an interview with AgFunderNews published in June, Upside chief operating officer Amy Chen confirmed that its whole-cut chicken product was still being made in 2-liter roller bottles. Upside’s website—which includes a page dedicated to explaining the production of its chicken—makes no mention of roller bottles.

“We are already producing whole cuts, whole tissue products. Rather than producing cells in a suspension or in a slurry, we are producing whole tissues directly out of the cultivator,” former senior vice president of operations Steve Myrick told Meat+Poultry magazine in April 2021. When Upside opened its pilot factory in Emeryville in November 2021, CEO Uma Valeti also touted its ability to make all forms of meat. “Here, you can produce any kind of meat, poultry, or seafood that you can imagine. Including both ground and whole cuts of meat,” he said at the opening ceremony, in a video livestreamed from the facility.

But former and current employees say that these claims exaggerate Upside’s technical capabilities. Internally, employees would joke that the startup could be the next Theranos—the blood testing startup that imploded spectacularly, ending with its founder, Elizabeth Holmes, being convicted of fraud. “It was a running joke: ‘Are we the next Theranos?’” says one former employee. “I don’t think it necessarily means that they are the next Theranos,” the former employee adds. “No one is dying. People are being lied to, but no one is going to die. Ideally.” Another former employee confirmed that staffers at the company would make jokes comparing Upside to Theranos.

Now, Upside seems to be putting its whole-cut fillets on the back burner. On September 14, Upside announced plans to build a 187,000-square-foot facility in Glenview, Illinois, that will initially be devoted to creating nugget-like ground-chicken products—a process for which Upside has yet to receive a green light from regulators. Announcing the new facility, Upside said that, once complete, it would not produce the “whole-textured” chicken it’s been serving, though the company aims to do so “in the future.”

The public tasting at Bar Crenn was supposed to signal that the era of lab-grown meat had finally arrived. Instead, sources claim, Upside has struggled with technical setbacks, while projecting an image of having solved the key scientific challenge of scaling up the production of whole cuts of meat. The revelations raise questions about exactly how much cultured meat companies, after nearly $3 billion in investment over the past seven years, have accomplished—and whether certain types of cultivated meat products can ever be commercially viable.

Meaty Claims

For a nascent industry, the cultured-meat startup scene is crowded. Every company has its own spin on the technology—a way to set itself apart from competitors. Some are opting to replicate high-end products like sushi-grade salmon. Others are going more exotic, with strange riffs on mammoth meat. California-based SciFi Foods is leaning into the weirdness of it all, experimenting with beef burgers that have as little as 5 percent animal cells mixed with plant-based ingredients.

Upside’s ability to produce whole cuts of meat is a point of distinction from its competitors—an apparent technological head start that has given Upside an edge when it comes to raising funds. The startup has raised more than $600 million in funding since 2016—attracting more than a fifth of all the capital raised by cultured meat companies up to the end of 2022, according to data from PitchBook and the Good Food Institute. Its list of investors includes SoftBank, Cargill, Richard Branson, and the Abu Dhabi Growth Fund.

Upside’s whole-cut chicken. Photograph: Christie Hemm Klok

There are, roughly speaking, two ways of turning animal cells into a cultivated meat product. The easiest and cheapest is to grow cells in suspension, which means mixing free-floating cells in a bioreactor with liquid feed and waiting until those cells have divided and matured. These suspension cells can then be harvested as a meat slurry and processed into ground-meat products like hot dogs and chicken nuggets. Adding plant-based ingredients and processing the resulting mixture can help mimic the texture of whole-cut meats, but to really nail that mouthfeel, companies will probably need to go beyond suspension cells, says David Kaplan, executive director of the Tufts University Center for Cellular Agriculture in Boston.

The other option is to find a way to make cells knit together and form sheets of tissue as they grow. Stacking and pressing these sheets together can result in a texture that is closer to a chicken breast than a chicken nugget.

This is the process that Upside presented to the US Food and Drug Administration in October 2021. In November 2022 the FDA responded to Upside with a “no further questions” letter about the safety of its process—the first issued in the US to a cultivated meat company and a major milestone on the road to getting its product approved. It is this FDA-recognized process that Upside implies its Emeryville factory is capable of deploying at scale—using multiple custom-made, 500-liter bioreactors.

But former and current Upside employees say the company struggled to use its large tanks to produce sheets of tissue that could be made into whole cuts of meat. One former employee says that between the factory opening in November 2021 and the summer of 2022, they saw dozens of attempts to use the bioreactors to produce sheets of tissue, but they rarely resulted in usable meat. At times, production runs were ruined by contamination that meant the meat was unsuitable for turning into a product, the former employee says.

Former Upside employees describe how batches of meat growing in the custom-made bioreactors would frequently be ruined by contamination and have to be incinerated. “Once they had any indication it was contaminating, they would try to just stop the run, get the cells, and get any results out of it that they could,” says a former employee with knowledge of the process.

A current Upside employee who describes the same problems says the proprietary tissue cultivators are no longer being used to brew sheets of cells, despite their prominent placement within the facility—and also are not brewing tissue for the Bar Crenn partnership. The issues with the bioreactors, key components of Upside’s intellectual property stack, have so far proven arduous. And though Upside is still working on designs for a replacement model, the current employee says, it has decided to go ahead with the limited launch of its whole-cut chicken even as it attempts to develop a scalable process.

“One day people are going to find out that none of those things work,” the employee says of the tissue cultivators. But Upside staffers continue to show off the reactors to guests on tours of the facility, suggesting that they are part of a functioning process, the employee says. “It’s like, cute story, y’all,” the current employee adds. Valeti pointed the bioreactors out to WIRED on the factory tour in May 2022. “These are a special set of cultivators, where we can do ground meats and also whole cuts,” he said at the time.

An industry insider with firsthand knowledge of the situation, but who has asked to remain anonymous out of fear of professional repercussions, also confirms that the custom-made tissue cultivators have recently been sitting empty, despite their prominent placement in Upside’s facility.

Inside Upside’s Emeryville factory. Photograph: Christie Hemm Klok

With the main bioreactors unsuitable for making whole cuts, former and current employees describe how Upside’s production of chicken fillets instead hinges on the laborious process of growing thin layers of cells in 2-liter-capacity plastic flasks commonly called roller bottles. One employee who worked on roller bottle production and left the company in 2022 described to WIRED the process of growing the tissue sheets. Lab technicians would start by coating the inside of bottles with porcine gelatin to help cells stick to the flasks’ surface. They’d later fill the bottles with a small amount of chicken cells and add growth media—a rich broth of hormones, sugars, and other nutrients.

After the bottles had spent around seven days being gently rolled back and forth in a heated incubation cabinet, technicians would scrape a thin layer of cells from the bottle by hand using a tiny squeegee. (Upside disputes this characterization. “It is not a squeegee. It is a custom-made spatula,” says interim head of communications Melissa Musiker.)

These thin layers, described by one source as resembling a “chicken fruit roll-up,” can then be stacked and molded together to mimic a whole cut of chicken. A current employee has confirmed that this process—so far unseen by the public and set apart from the main factory floor—is still being used to create the chicken served at Bar Crenn. “To my knowledge, roller bottles have been their bread and butter,” says one former employee. “They had a lot of confidence in roller bottles.”

WIRED sent a detailed list of questions to Upside based on our reporting. In an emailed statement, Chen, the company’s chief operating officer, claimed that our reporting contained a number of factual inaccuracies but did not directly address any of the points raised. Chen went on to say that Upside’s current process for producing chicken fillets was not one the company intends “to scale in its current form.”

“All breakthrough innovations and transformative technologies, including scaling our cultivated meat products, will take time,” Chen wrote in the emailed statement. “Cultivated meat is no exception. Not every research avenue, cultivator, or idea we explore will materialize exactly as we expected.”

Small Yields, Huge Costs

The current employee estimates that each roller bottle can produce between 2 and 3 grams of usable tissue. An average chicken breast weighs about 170 grams. Industry experts consulted by WIRED agreed that each 2-liter roller bottle would likely yield just a few grams of meat—many orders of magnitude less than one would expect from a large, working bioreactor.

One expert likens this approach to making food in a home kitchen. “You can’t build an industry from roller bottles,” says Ricardo San Martin, director of the Alt: Meat Lab at UC Berkeley. “No one produces things at scale in roller bottles. It’s a lab technique.”

While the resulting meat might make a tasty meal for a handful of people, that doesn’t solve cultivated meat’s two major problems: cost and scale. “It’s probably a useful intermediate-level solution to get some product out and get some feedback while new innovation and further scaling is being worked on behind the scenes,” said Kaplan when WIRED described to him the roller bottle process being used by Upside.

Upside has repeatedly suggested it’s ready to mass-produce muscle-like cuts in press interviews and press releases, including in November 2021when it claimed its Emeryville facility could “produce ground or whole cuts of meat, poultry, or seafood” at a rate of 50,000 pounds per year. As long as it’s dependent on roller bottles, though, the startup will struggle to serve up even scanty portions of cultivated meat.

Bar Crenn currently offers diners a 1-ounce portion of Upside’s chicken as part of a six-course, $150 meal, which, according to Vox, is available to 16 diners across one weekend each month. Sixteen ounces of salable meat per month is a long way from the more than 4,000 pounds per month that Upside says its factory is capable of producing. Previous reporting by the The Wall Street Journal in April also confirmed Upside’s use of roller bottles, though this was before Upside’s chicken fillets were made available to diners at Bar Crenn. The restaurant did not respond to a request for comment.

A sensory room where Upside’s chicken is evaluated by expert tasting panels. Photograph: Christie Hemm Klok

But roller bottles aren’t just problematic due to their miniscule output. They’re also expensive and wasteful, according to numerous biopharma and cultivated-meat industry sources. Each flask is made from sterile, single-use plastic, which must be discarded or recycled once it has generated its 2 to 3 grams of tissue. Given that an ordinary 2-liter plastic drink bottle can easily weigh 30 grams, Upside’s current production method likely produces plastic waste more than 10 times faster than it makes meat. “Can you imagine how much landfill waste you’re actually generating?” says one former employee, who was disturbed by the volume of single-use plastic being used at Upside.

Roller bottles also require a lot of tending by human beings, says David Humbird, an independent analyst and bioprocess scale-up expert and author of a comprehensive economic assessment of cultivated meat. Before it can produce its 2 to 3 grams of tissue, each bottle will need to be individually prepared, filled with cells, then periodically opened and filled with new media over the course of several days—a painstaking process that staffers must repeat under sterile conditions with hundreds of bottles before finished cell sheets can be carefully squeegeed out. It’s an approach that requires many hours of work by trained scientists to yield a tiny serving of food.

There are two distinct goals in scaling up, Humbird says: increasing the volume of what you make and how efficiently you make it. He argues that roller bottles fail in both respects—and that diners eating Upside’s chicken at a restaurant should be made aware of the “thousands of dollars of labor” that went into preparing the cultivated meat for each night’s meal service, regardless of what they pay at the end of the meal.

Other breakthroughs touted by Upside do not appear in its launch product. In December 2021, the company announced it had created a cell-growth liquid that is completely free of animal components. At the time, the company noted this “milestone” as a “crucial step” to making humane, cost-effective cultivated meat. Avoiding animal components—apart from the initial cell sample—is a major goal in an industry driven by animal welfare as much as environmental concerns.

But the chicken served at Bar Crenn uses multiple animal-derived ingredients. The roller bottles in which the tissues are grown are coated with porcine gelatin to help the cells stick to the plastic surface. An FDA safety dossier submitted by Upside also confirms that it might use bovine serum to grow its cells. The current employee also said that bovine serum—derived from the blood of adult cattle—is being used in the roller bottle process for Bar Crenn.

“This isn’t about lying. This isn’t about trying to trick people,” says a former senior Upside employee who left in 2022. “It’s a very specific PR dance where we need people to continue to believe in this industry and its potential, and to do that we need to share some of these milestones.” The former employee adds that the drive by senior leadership to constantly be the first to win industry races might jeopardize Upside’s success in the long term. “The biggest fear was more that we are constantly building the airplane while we fly it,” they say. “That could work out well, but would it be more prudent for us to pause flying and just continue building it and make sure it works as intended?”

Regulatory Headaches

Upside’s early commitment to whole cuts of chicken may come back to haunt it. In November 2022, it became the first cultured meat company to clear a premarket consultation from the FDA, in which the agency evaluated the safety of Upside’s production process and the meat it produced—a major step on the road to selling its products. “It’s a big, big step. It’s the biggest moment of maximal validation we’ve had to date, and I couldn’t be happier,” CEO Uma Valeti told WIRED at the time.

But the consultation applied only to chicken made using Upside’s current, problematic processes: sheets of tissue produced in roller bottles and tissue cultivators. If the company wants to switch to a simpler production method more suited to making ground products cheaply and at scale, it will require further clearance from the FDA. FDA press officer Veronika Pfaeffle confirmed that its previous consultation of Upside’s process applied only to cells harvested in the form of thin sheets of cells.

That puts Upside in a bind. It can take up to a year for the FDA to grant an approval after a safety dossier has been submitted, says David Tonucci, a regulatory affairs expert who was formerly the regulatory and toxicology vice president for the cultivated-meat startup SciFi Foods. “That’s only because the FDA reserves the right to come up with additional questions, which they almost always do,” he says. “And then just to go through the regulatory internal process will take nine to 12 months. That’s not to say that any application is good or bad—that’s just the process.”

Upside’s competitor, Eat Just, already has the regulator’s green light to produce cultivated chicken using more traditional suspension reactors. Its chicken is currently being served at the restaurant China Chilcano in Washington, DC.

Despite this lack of approval for a suspension-cell product, statements made by Upside executives have implied that the company is ready to produce large quantities of meat. “We’ll be able to produce 50,000 to 75,000 pounds of meat every year, right away,” Valeti is quoted as saying in an article published by CBS News on July 9.

Upside’s whole-cut chicken (right), alongside two other products from its portfolio. Photograph: Christie Hemm Klok

David Kaplan at Tufts University doesn’t see this as a major problem for Upside. “It wouldn’t surprise me if a large company like that is going to pivot to suspension culture and suspension meats,” he says. If the company can get out there with ground meat products, it might buy more time to figure out a way to grow whole cuts that doesn’t rely on problematic roller bottles.

No company has yet solved the problem of scaling cultivated meat, says Johnny Ream, a partner at venture capital firm Stray Dog Capital, which invested in Upside in 2016. Ream noted that as an early investor he had limited access to information about the company but that Upside had still made major achievements. “We just need to remember that just being able to sell in the United States, getting approval from the USDA [United States Department of Agriculture], these are great things. They’re good milestones, but we’re still in early innings here,” he says.

In the emailed statement, Upside’s Chen said the company was “continuing to work on processes” to improve the taste, texture, scalability, and cost of its products. “We have already made significant progress on this front—including leapfrogging some of our earlier generation technology,” Chen added.

But the fact that Upside is struggling to create whole cuts of meat did not come as a surprise to investors outside the company.“ The reason that so few companies are going after [whole cuts] is because it’s really fucking hard,” says a prominent investor in the field of cultivated meat who has been pitched by Upside Foods but has not invested in the company. “It’s just a level of technology that doesn’t need to be applied to this industry at this point in time, and I just don’t see how that could be scalable right now.”

So why did Upside try? The pursuit of whole cuts of cultivated meat might have more to do with generating enthusiasm for the technology than proving its commercial viability, says Gabriel Ruimy, cofounder of investment firm Bloom8, who is familiar with the cultivated-meat industry. “This makes noise. When you make noise, you create a visible future for the technology and attract investors; when you attract investors, you bring dollars, and with dollars you can improve your technology and bring forward the whole market with you.”

This is the challenge facing Upside and the cultivated-meat industry more widely. Can it generate the enthusiasm—and money—needed to buy enough time to solve the technological hurdles that remain?

Despite the problems at its pilot facility, Upside is pressing ahead with plans to spend more than $130 million in the US Midwest, with efforts centered on the newly announced, commercial-scale plant in Glenview. Upside will call the facility Rubicon—or the point of no return. That plant, once operational, will be able to produce millions of pounds of ground, cultivated chicken, according to the company. Upside did not provide a timeline for when the whole-cut products it’s serving at Bar Crenn would be available at scale, though it did recently post a job opening for a plant manager based near Chicago or southeast Wisconsin.